Tunadorable

fixing flash-attention kernel 43:09
fixing flash-attention kernel Tunadorable 54 6 часов назад
Building GPU kernel tutorials for dropout & layernorm - live 2025.2.21 20:06
Building GPU kernel tutorials for dropout & layernorm - live 2025.2.21 Tunadorable 38 7 часов назад
LayerNorm | Triton GPU Kernels 101 Lesson #8 1:22:39
LayerNorm | Triton GPU Kernels 101 Lesson #8 Tunadorable 113 3 дня назад
Matmul | Triton GPU Kernels 101 Lesson #6 1:20:14
Matmul | Triton GPU Kernels 101 Lesson #6 Tunadorable 164 5 дней назад
Dropout | Triton GPU Kernels 101 Lesson #7 11:41
Dropout | Triton GPU Kernels 101 Lesson #7 Tunadorable 81 5 дней назад
Fused Softmax | Triton GPU Kernels 101 Lesson #5 1:05:13
Fused Softmax | Triton GPU Kernels 101 Lesson #5 Tunadorable 79 5 дней назад
Vector addition | Triton GPU Kernels 101 Lesson #4 41:07
Vector addition | Triton GPU Kernels 101 Lesson #4 Tunadorable 135 5 дней назад
How to use a cloud GPU | Triton GPU Kernels 101 Lesson #3 18:22
How to use a cloud GPU | Triton GPU Kernels 101 Lesson #3 Tunadorable 198 5 дней назад
GPU Architecture Basics | Triton GPU Kernels 101 Lesson #2 15:10
GPU Architecture Basics | Triton GPU Kernels 101 Lesson #2 Tunadorable 183 5 дней назад
Building GPU kernel tutorials for dropout & layernorm - live 2025.2.21 20:06
Building GPU kernel tutorials for dropout & layernorm - live 2025.2.21 Tunadorable 38 7 часов назад
Have we been doing LLM inference wrong the whole time?!?! 26:53
Have we been doing LLM inference wrong the whole time?!?! Tunadorable 6,258 1 месяц назад
Normalizing GPT on the unit-hypersphere (WITH CODE) 1:05:35
Normalizing GPT on the unit-hypersphere (WITH CODE) Tunadorable 4,420 3 месяца назад
The Structured Task Hypothesis 10:40
The Structured Task Hypothesis Tunadorable 1,826 7 месяцев назад
Why Does Diffusion Work Better than Auto-Regression? 20:18
Why Does Diffusion Work Better than Auto-Regression? Algorithmic Simplicity 485,400 1 год назад
Does AI have any chance predicting chaotic systems? 20:13
Does AI have any chance predicting chaotic systems? Tunadorable 4,133 5 месяцев назад
Conversational Swarm Intelligence with Dr. Louis Rosenberg 1:04:34
Conversational Swarm Intelligence with Dr. Louis Rosenberg Tunadorable 1,045 8 дней назад
purposely "pre-caching" features or inadvertently leaving "breadcrumbs" for future timesteps? 22:03
purposely "pre-caching" features or inadvertently leaving "breadcrumbs" for future timesteps? Tunadorable 2,328 6 месяцев назад
SORA DÉTRÔNÉ ? Le CHOC des IA vidéo que personne n'attendait ! 19:59
SORA DÉTRÔNÉ ? Le CHOC des IA vidéo que personne n'attendait ! Johan : Solutions Digitales 3,441 1 день назад
Claude Almost Bankrupt Me... 16:18
Claude Almost Bankrupt Me... Theo - t3․gg 91,507 55 лет назад
Do we really need to use every single transformer layer? 21:43
Do we really need to use every single transformer layer? Tunadorable 2,290 5 месяцев назад
Sigma-GPTs: A New Approach to Autoregressive Models 23:47
Sigma-GPTs: A New Approach to Autoregressive Models Tunadorable 2,888 7 месяцев назад
Accelerated Training by Amplifying Slow Gradients 26:29
Accelerated Training by Amplifying Slow Gradients Tunadorable 32,244 8 месяцев назад
Are our perceptual systems structured to view the world truthfully? 31:57
Are our perceptual systems structured to view the world truthfully? Tunadorable 2,766 6 месяцев назад
Curating December's best new AI papers from ArXiv 27:52
Curating December's best new AI papers from ArXiv Tunadorable 1,734 2 месяца назад
Rambling about GPT architecture edit ideas 42:13
Rambling about GPT architecture edit ideas Tunadorable 1,460 5 месяцев назад
How to make neural networks better at learning new things 33:33
How to make neural networks better at learning new things Tunadorable 1,943 55 лет назад
What would it mean for an AI to "understand"? 1:00:28
What would it mean for an AI to "understand"? Tunadorable 3,749 6 месяцев назад
Let this method tune hyper-parameters for you! 19:02
Let this method tune hyper-parameters for you! Tunadorable 2,122 1 месяц назад
A new way to compare high dimensional vectors 9:11
A new way to compare high dimensional vectors Tunadorable 11,399 6 месяцев назад
Creating new tokens out of internal representations 16:21
Creating new tokens out of internal representations Tunadorable 2,685 2 месяца назад
Skimming hella AI paper abstracts - Nov 5, 2024 51:50
Skimming hella AI paper abstracts - Nov 5, 2024 Tunadorable 1,607 3 месяца назад
channel update 13:05
channel update Tunadorable 1,644 55 лет назад
Models inside models inside models 26:40
Models inside models inside models Tunadorable 2,347 5 месяцев назад
Some training tokens are more valuable than others 8:49
Some training tokens are more valuable than others Tunadorable 1,396 6 месяцев назад
Theoretical physics of next token prediction in LLMs 9:13
Theoretical physics of next token prediction in LLMs Tunadorable 2,127 3 месяца назад
How diffusion modeling can revolutionize evolutionary algorithms 21:30
How diffusion modeling can revolutionize evolutionary algorithms Tunadorable 5,673 1 месяц назад
Skimming hella new AI paper abstracts - January 2025 39:21
Skimming hella new AI paper abstracts - January 2025 Tunadorable 1,882 2 месяца назад

Сейчас ищут